Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference, MSIEID 2023, December 8–10, 2023, Guangzhou, China

Research Article

Design and Implementation of Speech Recognition System Based on Artificial Intelligence

Download83 downloads
  • @INPROCEEDINGS{10.4108/eai.8-12-2023.2344688,
        author={Yu  Li},
        title={Design and Implementation of Speech Recognition System Based on Artificial Intelligence},
        proceedings={Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference, MSIEID 2023, December 8--10, 2023, Guangzhou, China},
        publisher={EAI},
        proceedings_a={MSIEID},
        year={2024},
        month={4},
        keywords={artificial intelligence ai; voice recognition; on-board alarm system; system framework},
        doi={10.4108/eai.8-12-2023.2344688}
    }
    
  • Yu Li
    Year: 2024
    Design and Implementation of Speech Recognition System Based on Artificial Intelligence
    MSIEID
    EAI
    DOI: 10.4108/eai.8-12-2023.2344688
Yu Li1,*
  • 1: School of International Relations
*Contact email: 917586724@qq.com

Abstract

Speech recognition technology based on artificial intelligence AI technology has gradually been widely used under the background of the continuous development of current data processing technology, and belongs to the branch of biometrics technology. AI-based speech recognition technology can be applied to biometric recognition and matching in different fields, providing technical support for the intelligent development of various industries and fields. In car intelligent voice recognition system, for example, for the car voice recognition system speech recognition real-time and accuracy requires high characteristics, the article adopts multiple pickup as car voice system sound acquisition hardware system, the voice data pretreatment after feature extraction, in artificial intelligence AI technology and sound biometric matching model and alarm sensitive words. In addition, the sensitive word recognition transmission system is connected to the cloud platform of the public security alarm system to realize intelligent voice recognition alarm. In addition, the dialect speech library of different regions was established, 60 experimental subjects from different regions were selected, and 5 alarm sensitive words were spoken in dialect for the experiment.